Emergence Economics

7/28/2019 Emergence Economics

1/77

MAY 2008 DRAFT

Please do not quote or cite without permission.

The New Emergence Economicsof Innovation and Growth,

And What It Means forCommunications Policy

May 2008

Richard Whitt and Stephen Schultze

It is not the strongest of the species that survives, nor the most intelligent, butthe one most responsive to change. ~Charles Darwin, The Origin of Species


2/77

The New Emergence Economics of Innovation and Growth Page i

INTRODUCTION......................................................................................... 1

EMERGENT ECONOMY: OVERTHROWING THE OLD REGIME.................4

INTRODUCING EMERGENCE ECONOMICS ........................................................................................................ 4

The Nature of Markets: Complex Cascades........................................................................................... 5

The Nature of Competition: Imperfect Incentives ................................................................................. 7

The Nature of People: Behavioral Beings............................................................................................... 8

The Nature of Analysis: Mismatched Models ........................................................................................ 9

PRESENTING AROUGH FORMULAFOREMERGENCE...................................................................................... 10

Agents..................................................................................................................................................... 11

Networks ................................................................................................................................................14

Evolution................................................................................................................................................17

Emergence of Networks and Growth.................................................................................................... 22

NETWORKED ECONOMY: THE INTERNET AS THE ULTIMATEEMERGENT PLATFORM...........................................................................24

THE NETS ORIGINS ..................................................................................................................................... 25

Overlooked Components: The Social, Economic, and Legal Backdrop.............................................. 25

An Unlikely Birth .................................................................................................................................. 26

THE NETSARCHITECTURE ..........................................................................................................................28

The Law of Code: Modularity .............................................................................................................. 28

Smart Edges: End-to-End.................................................................................................................... 29

A Network of Networks: Interconnection .............................................................................................31

Agnostic Protocols: IP...........................................................................................................................31

THE END RESULT:AVIRTUOUS FEEDBACKNETWORK .............................................................................. 32

GROWTH ECONOMY: THE EMERGENCE OF IDEAS, INNOVATION, ANDNET EFFECTS........................................................................................ 33

THE NATURE OF IDEAS AND INNOVATION..................................................................................................... 34

Ideas ...................................................................................................................................................... 35

Innovation............................................................................................................................................. 36ECONOMIC GROWTH.................................................................................................................................... 39

The Road to Romer...............................................................................................................................40

Enter New Growth Theory.....................................................................................................................41

Implications For Technological Change............................................................................................... 43

THE EXTERNAL IS IN: NET EFFECTS........................................................................................................... 44

Innovation Spillovers.................................................................................................45

Richard Whitt and Stephen Schultze May 2008


3/77

The New Emergence Economics of Innovation and Growth Page ii


Peer Production .................................................................................................................................... 46

The Social Layer................................................................................................................................. 47

ANEW VISION OF WEALTH............................................................................................................................48

POLITICAL ECONOMY: EMERGING IMPLICATIONS FOR U.S.COMMUNICATIONS POLICY....................................................................49

THE OVERALL ROLE OF GOVERNMENT ......................................................................................................... 50

The Co-Evolution of Markets and Governments.................................................................................. 50

The Policymaker As Adaptive Economic Agent ................................................................................... 52

ACOMMUNICATIONS POLICYAPPROACH FORINNOVATIONAND GROWTH .................................................... 53

Why Communications Policy? .............................................................................................................. 53

Defining Our Mission............................................................................................................................58

APPLYING THE FRAMEWORK:ENABLING WITHOUT DICTATING ......................................................................64

Not Dictating the Evolutionary Process ............................................................................................... 64Enabling the Econosphere................................................................................................................. 67

CONCLUSION ........................................................................................... 74


4/77

The New Emergence Economicsof Growth and Innovation,

And What It Means for

Communications PolicyMay 2008

Richard Whitt and Stephen Schultze

INTRODUCTION

Your college economics professor probably lied to you. Or fibbed, or unwittingly gave you falseinformation, or was merely confused. That is the startling yet inescapable conclusion after onereviews the latest empirical and analytical work in the schools we will refer to collectively asEmergence Economics. The hoary Old School Economics still teaches us, for example, thatthe market is linear and always seeks equilibrium, that economic actors are perfectly rational,with perfect knowledge of themselves and the marketplace, that production is generated only bycapital markets or government subsidy, that growth is exogenous, and the whole of the economicsystem is always equal to the sum of its parts. It turns out that every one of these key

assumptions is plain wrong. While the ramifications for the study and practice of economicscannot be overstated, the implications for the public policy world that has rested for so long onthose assumptions cannot be overlooked.

Todays discussions about national communications policy often seem to be rooted to the past, inthe form of economic and technology assumptions that more or less ended in the 1960s. As itturns out, the rise of new economic thinking, along with new technology platforms culminatingin the Internet, together directly challenge many of those chief assumptions. In the authorsview, now is an appropriate time to articulate the fundamental economic and technology tenetsthat should inform our nations communications and information policies, and to beginsuggesting some ways those policies should be recast in their light.

Twenty years ago, no one would have anticipated the Internet as we know it today. Consumer-grade computer modems were still in their infancy, and the few dial-up online communities thatexisted were a far cry from the globally connected network of networks that now pervades somuch of what we do. In many ways, the Internet was a happy accident. What started out asisolated islands universities, bulletin board systems, commercial services linked together and

Mr. Whitt currently is Washington Telecom and Media Counsel at Google Inc. Mr. Schultze served as a policyintern with Google.


5/77

The New Emergence Economics of Innovation and Growth Page 2

grew as the result of the actions of millions of unaffiliated people. The underlying softwareprotocols opened up the ability to interact and speak freely across thousands of interconnectednetworks. The growth of the Internet, and the online and offline economy it facilitates, wasbeyond even the wildest predictions. In many ways, the Internet is what happened while wewere busy planning something else.

Even with the benefit of hindsight, we do not fully understand what led to this success.Advances in computer technology, including digitization of information, dramatic increases incomputing power, and concomitant declines in the cost of computer storage, were necessaryconditions, but they also could have facilitated a variety of other outcomes. To comprehend howthe Internet developed into a thriving marketplace of innovation and economic growth, manyturn understandably to the field of economics. Unfortunately, traditional economics has little tosay about this leading test case.

Fortunately, two fundamental observations drawn from Emergence Economics lead us toward amore helpful analytic framework.

First, we live in a network economy, formed bottom-up by interactions between people in ahighly connected marketplace. Some basic rules govern these interactions, but for the most partthe system emerges freely and unpredictably. Economic actors become nodes, and the structureof the market evolves based on their collective practices. Whereas traditional economics moreoften than not attempts to engineer statically these relationships, Emergence Economicsrecognizes that such an abstraction fails to represent accurately a far messier reality. Instead, thenetwork economy thrives when there is space for experimental evolution, in which new ideasemerge and technology constantly is refined.

Second, we live in a growth economy in which the chief currency is ideas, and the mechanismfor growth is innovation. While traditional economics tells us that productivity comes simply

from adding more capital, or generating greater efficiency, Emergence Economics emphasizesways in which technologies, broadly defined, transform the means of production. The advent ofnew technologies can create better recipes for economic growth. Whats more, thesetechnologies do not emerge inexplicably from outside the system, but instead are the product ofeconomic actors working within (indeed, comprising) the system according to diverse motives.In addition, growth is not limited by physical goods, but is enabled by the transmission,reproduction, and improvement of ideas. An economic framework that actually recognizes thisdynamic can establish a foundation for even greater growth in the future.

This networked growth economy differs greatly from some traditional economic models thatposit static, linear forms of growth. Economists often use the phrase virtuous circle to describe

systems that contain positive feedback loops in which their outputs cycle back into their inputs.Rather than moving toward equilibrium, these economies are self-reinforcing and have thepotential to multiply their effects in unexpected ways generating both positive and negativefeedback. The virtuous circle appears in the form of technological innovation that facilitatesfuture technological innovation and drives network effects where each new user adds benefitto the rest of the system. Emergent economies combine these dynamics, harnessing thediscoveries of others in such a way that the system as a whole grows far more efficiently than ifit were a disconnected set of actors. The interdependent virtuous circles can become part of acomplexvirtuous feedback network.



6/77



And so what of the proper role of the government policymaker the legislator, the regulator,the reviewing judge in the face of an innovation-fueled, network-connected, emergenteconomy? There is little doubt that the policy environment needs to catch up to newer economicthinking. As we shall see, the lessons here are many; at bottom, they point to caution, and evenoutright skepticism, about becoming a more active force in the market, but always tempered by

optimism. The tools of government, when employed sparingly, carefully, and deliberately, andin the right context, can successfully facilitate a more optimal environment for new ideas,economic growth, and human freedom. However, this bottom-up regulatory approach requiresan appreciation for, and understanding of, topics like network-based dynamics, and theconditions that are most likely to foster productive tipping points.

This paper will introduce the rough formula for Emergence Economics namely, thatindividual agents, acting through interconnected networks, engage in the evolutionary marketprocesses of differentiating, selecting, and amplifying certain business plans and technologies,which in turn generates a host of positive emergent economic phenomena. This formula isfueled by the latest findings from physics, biology, psychology, cognitive neuroscience, andplain common sense. The Internet then will be discussed as a notable and perhaps uniqueproduct of market and non-market forces, as a modular infrastructure, and as a platform forbroad-based innovation. Next, the paper will turn to some key emergent phenomena, includingideas, innovation, economic growth, and what we call Net effects. Finally, we will bring theseeconomic and technological elements to bear in the world of communications policy, where aproposed new framework separates out the virtues of tinkering with market gaps and inputs,versus the vices of tampering with evolutionary processes and outcomes.

We have a few important caveats to relay at the outset. First, the literature in this field is broadand deep in some places, scanty and unfinished in others. Nonetheless, as Michael Shermer,rightly declares, economics is undergoing the most dynamic revolution since Adam Smith,with rich indisciplinary hybrids emerging to breathe new life into an old science.1 This paper

necessarily presents an overview of what so far is known, or surmised, or even guessed at, but itdoes not purport to replace the foundational work of many others. Nor do we seek to discreditthe world of modern economic thinking; indeed, many of the intellectual trends we discuss hereslowly but surely are being incorporated into more mainstream schools. In particular, we willlook to Joseph Schumpeter and Friedrich Hayek as two towering transitional figures between thedisparate phases of Traditional Economics and Emerging Economics. The chief concern is that,while Traditional Economics may be adjusting itself to new ways of thinking, manypolicymakers still cling to the basic assumptions that underpin older forms of economic theory.As a result, our objective here is comparatively humble, yet practical: to condense what currentlycomprises this sprawling body of new work in a way that can aid policymakers and others inanalyzing policy issues, particularly in the communications realm. This particular dynamic

revolution needs to be televised.

In addition, all the coming discussion of markets and systems and properties should not letus lose sight of the common human element. Whatever happens in any agent-constructed spacesuch as economics should be taken to reflect humanity, in all its breadth and depth. Words andconcepts are not the things they describe; they merely serve as organizing principles to try tomake sense of a seemingly disorganized world. Ironically, economics, technology, and law

1 Michael Shermer, The Mind of the Market (2007), at xviii.


7/77



together which form the three-way intersection for this paper -- often are perceived as sterile,artificial, and even tedious fields. We would argue instead that they should be seen as the flesh-and-blood instantiation of ordinary humans participating in the world. Amidst the vernacular,we should never forget the people behind it all.

Moreover, it must be stressed that all economic models inherently are wrong, to one degree oranother. As mere abstractions of reality, they inevitably miss at least some of the nuance andsinew of the world that each of us inhabits. While the concepts explored here should not beconfused with exacting correlates to the real world, nonetheless, they do provide an importantcorrective to what has gone before.

Finally, we must point out that neither author of this paper is an economist, at least (yet) bytrade. Perhaps some will dismiss what follows solely for that reason; afterall, the saying goes,the only thing more dangerous than an economist is an amateur economist.2 Still, the world ofeconomic theory should not only be available to, and articulated by, a cloistered few.3 AsTraditional Economics gradually, and we think inevitably, makes way to new forms of economicthinking, we can only hope that more of us are able to join in the evolving conversations.

EMERGENT ECONOMY: OVERTHROWING THE OLD REGIME

INTRODUCING EMERGENCE ECONOMICS

So what exactly is this new economics? The term Complexity Economics adopted by analyst

Eric Beinhocker describes a variety of different analytic and empirical approaches to theeconomy, with the binding principle of attempting faithfulness to the real world. In this paperwe will be combining Beinhockers impressive synthesis with emerging work in other areas,most notably network science, new growth theory, and behavioral economics. Other usefulcontributions include the origins of innovation, competition theory, Net effects like spilloversand social production, and Ormerods Iron Law of Failure. As a result, we thought it best toavoid confusion by using a broader umbrella phrase, Emergence Economics, to denote thismore wide-ranging set of emerging viewpoints.

In reality, this new economics draws from several long existent but submerged themes inmainstream economics. Phenomena thought to be minimally important, peripherally relevant, oroutside the scope of proper economic thinking altogether are being brought to the forefront. Inthe case of communications-based technology sectors, economist Hal Varian has noted that:

there are some forces that are particularly important in high-tech. [T]hese forcesare not "new"; indeed the forces at work in network industries in the 1990s are

2 Supposedly found in Bentleys Second Law of Economics.3 We also take heart from Hayeks statement that an economist who is only an economist cannot be a goodeconomist.


8/77



very similar to those that confronted the telephone and wireless industries in the1890s.4

Varian undoubtedly is correctmany of the various elements at play in todays economy havebeen there for many decades. Perhaps the best way to understand Emergence Economics is to

compare it to the more traditional economics handed down to us by the neoclassical theorists,and still the intellectual grounding for much of todays economic theory. Below we will lookbriefly at four aspects of traditional economic thought -- markets, competition, people, andanalysis and contrast them with the different perspectives brought by Emergence Economics.

One gating question first must be addressed, however: if Emergence Economics actually presenta more truthful version of the economic landscape, why have these ideas taken so long to beabsorbed into the social and political mainstream? Economist Paul Ormerod cites as likelyreasons sheer intellectual inertia and, until recently, a lack of sophisticated analytic tools.5Unfortunately, as we shall see, public policy tends to hew closely to the dictates of traditionaleconomic theory. In a small way, this paper seeks to offer a much-needed corrective to thatsituation.

The Nature of Markets: Complex Cascades

Neoclassical theory states that the economy is a static equilibrium system, existing at rest, andmoving from equilibrium point to point as it seeks balance. Under this view, the economyliterally is a closed equilibrium system.6 Neoclassical theory also sees an invisible hand atwork in competitive markets. In a free market economy, the thinking goes, a natural restingpoint is reached, where supply equals demand, resources are put to their most efficient use, andthe welfare of society is optimal. Such a market is deemed optimally efficient. The businesscycle is just that: a regular, periodic, and predictable movement between boom and recession.

Economic processes are dominated by dampening, negative feedback that keeps thingscontained. Any indeterminacy typically is assumed away by econometric models.7 The businesscycle is determined exogenously, by occasional shocks originating from outside the systemitself.8

By contrast, a central tenet of Emergence Economics is that the economy is a complex adaptivesystem (CAS), which is a subcategory of open systems.9 In a CAS, micro-level interactionslead to the emergence of macro-level patterns of behavior. As one example, financial markets

4 Varian, H. R., J . Farrell, et al. (2004). The Economics of Information Technology: An Introduction. Cambridge;New York, Cambridge University Press., at p. 3.5 Paul Ormerod, Why Most Things Fail, at ix. One author goes so far as to claim that traditional economist theoriesemploy an explicitly religious defense of the marketplace. Robert Nelson, Economics as Religion (2001). Anotherauthor notes that some supporters of traditional economics want so badly to believe in their models that,paradoxically, these economic formulas and models were symptoms of the very desires and emotions they weredesigned to eliminate. Mark Taylor, Confidence Games (2004), at 276.6 Mark Taylor, Confidence Games, at 239-40.7 Ormerod, Why Most Things Fail, at 79-80.8 Ormerod, Why Most Things Fail, at 191.9 W. Brian Arthur, The Economy As an Evolving Complex System (1997). Or, as Michael Shermer puts it,economies are complex systems that emerge out of the simple activities of people just trying to make a living andprovide for their children. Shermer, The Mind of the Market, at 5.


9/77



are far more turbulent, deceptive, and risky than previously thought, with prices leaping up anddown in a more or less concentrated fashion.10 Oftentimes, small, innocuous events can set offavalanches of change that are inexplicable in Traditional Economics, while large disturbancesultimately may have no lasting impact. A key aspect of any CAS is the inherent lack ofpredictability in its future operations, because they do not add up in a simple, linear way. Dips

and peaks in the economy refuse to recur in a predictable manner.

11

To quote 2002 Nobel Prizewinning economist Vernon Smith, We do not understand why markets work as they do.12

Further, the economy is not an equilibrium system at all, and will never reach a resting place, forsuch a state would equal death. Ormerod points out that there are numerous empirical andtheoretical bases to criticize the concept of general equilibrium.13 The real world also exhibitspositive feedback, or increasing returns; there are always new sources of positive feedback, sothere is no long run in the real world. Time is an important element of economic phenomena,one missing in general equilibrium theory.

The idea that the market is optimally efficient also runs headlong into reality. No less anauthority than Lawrence Summers has been quoted as saying that the efficient markethypothesis is the most remarkable error in the history of economic theory.14 Mark Taylor, anoted expert on complexity theory, agrees that the efficient market hypothesis and relatedtheories were wrong on virtually every count, so that the more carefully one ponders themarket, the more suspect the whole notion of efficiency becomes.15

Traditional Economics is riddled with these basic flaws for a good reason. It is widelyunderstood that neoclassical economic theory views the economy through the prism of thephysical sciences of the late 19th Century, particularly atomistic statistical mechanics.16 Themetaphor appropriated from the Industrial Revolution is the economy as a human-mademachine(like a steam engine), whose behavior is fixed, stable, predictable, and controllable.17 Thisperspective emphasizes rules executed by top-down hierarchies of relatively expert and impartial

officials who prize efficiency and consistency.18 By contrast, Emergence Economics views theeconomy through the prism of modern day physics and biology, with metaphors better suited tothe Information Revolution. Under that perspective, human society, of which the economy is asubset, is more like a living organism, a complex system characterized by constant change,evolution, and disequilibrium that percolate from the bottom up.19

10 Benoit Mandelbroit, The (Mis)behavior of Markets (2004), at 225-252.11 Philip Ball, Critical Mass (2002), at 189.12 Vernon Smith, CONSTRUCTIVIST AND ECOLOGICAL RATIONALITY IN ECONOMICS, Nobel PrizeLecture, December 8, 2002, at __.13 Ormerod, Why Most Things Fail, at 50-51.14 Lawrence H. Summers, Does the Stock Market Rationally Reflect Fundamental Values?,The J ournal ofFinance, Vol. 41, No. 3, Papers and Proceedings of the Forty-Fourth Annual Meeting of the America FinanceAssociation, New York, New York, December 28-30, 1985 (Jul., 1986), at __.15 Mark Taylor, Confidence Games, at 269, 273.16 SeeOrmerod, Why Most Things Fail, at 17-35; Ball, Critical Mass, at 204-06; Mark Taylor, Confidence Games,at 273.17 See generallyOrmerod, Butterfly Economics (1998); Axelrod and Cohen, Harnessing Complexity, at 28-31.18 Axelrod and Cohen, Harnessing Complexity, at 28-31.19 Ormerod, Butterfly Economics. Crucially, the neoclassical economists failed to include in their thinking theSecond Law of Thermodynamics, which describes entropy and the notion of an open system.


10/77



One illustrative example of the different approaches is the prices for goods and services.Traditionally one of the central tensions in economics is between producers (supply) andconsumers (demand). Supply and demand supposedly balance out precisely; there is no waste,and goods are distributed in a Pareto optimum way. Further, the price should always be setequal to marginal cost (or where additional revenues will exceed additional costs).20 Relatedly,

stock prices accurately reflect all available information at all times, and so should follow arandom walk, with no patterns and no clues for future prices.21

Emergence Economics challenges each of these assumptions.22 Supply rarely equals demand.Empirically, we often see a wide divergence in the prices of individual goods and services.Demand and cost curves are extremely difficult to know with any clarity.23 Uncertainty and lackof information shroud the future in doubt, which makes setting prices that much more difficult.

The Nature of Competition: Imperfect Incentives

The traditional view of economic competition is that perfect competition compels free markets

to allocate scarce resources in a manner so efficient that all our conflicting wants and needs areresolved in the most satisfactory manner possible.24 The watch phrase is consumersovereignty.25 Free markets are presumed to be both efficient and fair. The efficiency principlesays that under perfect competition, and with no market failures, free markets will squeeze asmany useful goods and services as possible out of the available resources (maximal output atminimal prices), and that anything that interferes with the price systems ability to do so is adetriment to social well-being.26 The concept that the supply of every traded good or service isprecisely equal to the demand for it at prevailing prices led to the related concept that theeconomy rests in perfect equilibrium.27

Now, however, the study of competition slowly is turning from a theoretical to an experimental

science. And the free market rarely, if ever, operates under conditions of perfectcompetition.28 Further, careful analysis shows that most firms are not pure maximizers of profitor utility, but also seek to attain other primary objectives such as attracting and retainingproductive workers.29 Common diversions to raising prices include predatory pricing,discriminatory pricing, price-fixing, attempts to monopolize, and unfair and deceptiveadvertising claims. In addition, the demonstrated phenomenon of sticky prices leads to higherthan necessary prices that refuse to descend to their theoretically sanctioned levels.30

20 [cites]21 Mark Taylor, Confidence Games, at 244-46.22 Robert Nelson calls these theories an economic tautology. Nelson, Economics as Religion, at 58-69.23 Ormerod, Why Most Things Fail, at 23-35; Ball, Critical Mass, at 254-55.24 Ormerod observes that the phrase perfect competition is yet another example of the terrifying ability ofeconomists to brand their central concepts so effectively. Ormerod, Why Most Things Fail, at 83-84.25 James Case, Competition (2007), at __.26 James Case, Competition, at 160-63.27 James Case, Competition, at 183-84; Ball, Critical Mass, at 191.28 Ball, Critical Mass, at 255.29 Ball, Critical Mass, at 257, 268-69.30 James Case, Competition, at __.


11/77



Traditional Economics also has failed to convey the reality of modern-day competition inanother fundamental way. The classical world view is founded on scarce material objects andtheir efficient allocation or as Paul Romer puts it, a finite quantity of things with which towork basically, the matter in the earths crust.31 Value comes from rearranging that matterinto a more valuable form. The physical world is characterized by diminishing returns, and

increasing cost per additional unit produced. Firms compete via prices in existing goods, andlaws were built around establishing property rights and ensuring no monopoly control. In theInformation Age, however, we rely increasingly on ideas as the recipes we use to rearrangematter to create more value and wealth.32 Because ideas are not scarce, the process ofdiscovering them does not suffer from diminishing returns. Indeed, increasing returns comesfrom both the shoulders-of-giants process (new ideas build on existing ideas, and then begetmore new ideas), and falling costs per unit (such as producing a software CD). Competition isfacilitated not by firms trying to drive prices, but firms seeking to capture market share throughnew products.

Joseph Schumpeter was an original prophet in this area. He saw claims about perfectcompetition as relatively unimportant; instead, what counts is competition from the newcommodity, the new technology, the new source of supply, the new type of organizationcompetition which strikes not at the marginsof the profits and the outputs of the existing firmsbut at their foundations and their very lives.33 Schumpeter also argued that some degree ofmonopoly is preferable to perfect competition, because supernormal profits are the temporaryfruits of innovation. Persistent competition from innovations creates a threat that disciplinesbefore it attacks, so that as soon as innovators cease to innovate, pricing power will desertthem.34

The Nature of People: Behavioral Beings

Under Traditional Economics, the pursuit of self-interest is a rational activity of individuals,based on utility (a measure of pleasure and pain). Paul Samuelson, a towering figure in 20thCentury economics, assumed that people are representative agents, both logical and consistentin their behaviors, and have perfect knowledge of all probabilities. Traditional economics alsoincludes assumptions about homo economicus possessing rational and errorless choice,presupposing perfect foresight, and foreknowledge free from uncertainty.35

Unlike the academic equations and assumptions that undergird traditional economics, behavioraleconomics is based on the actual discernible rules by which humans make everyday decisions.36In the real world, barriers to decision making almost always exist. Information is costly,

31 Joel Kurtzman, An Interview with Paul Romer, Strategy+Business, found at: http://www.strategy-business.com/press/16635507/9472.32 Id.33 Schumpeter, Socialism, Capitalism and Democracy, at 84.34 Id. Some modern-day students of imperfect competition do not necessarily agree with this point, as someoligopolies may cling to pricing power indefinitely. One salient example mentioned in the literature is Microsoft,which could have sold its Windows OS for $49, but instead chose a profit-maximizing price of $89. James Case,Competition, at 206. Indeed, these same economists observe that free markets long have tolerated all manner ofsupernormal profits, and such markets tend to evolve into tight oligopolies over time. Id.35James Case, Competition, at 293; Ball, Critical Mass, at 209-11.36 Ball, Critical Mass, at 213-14.


12/77



incomplete, and rapidly changing. At best, we employ bounded rationality, by makingdecisions in the face of obvious external and internal constraints.37 Ormerod observes that everyindividual decision involves massive complexity and defies the orderly application of the rationalcalculations of economic theory. Indeed, in the new economics, we not only address a specificproblem, we try to start from the outset with rules of behavior which have empirical support

rather than with rules that we believe a priori a rational agent ought to follow.

38

We willdiscuss this critical point in further detail in Part II.

Further, there is no such thing as self-interested individuals acting in isolation. Hayek showed usthat desirable outcomes are the joint product of both individual actions and the institutionalframework in which individuals operate.39 Social change isboth volatile and often inexplicable,as agents engage in clustering and herding behavior.40 Ample evidence shows that, inBeinhockers words, the interactions of millions of people, making decisions, engaging instrong reciprocal behavior, acting out their cultural norms, cooperating, competing, and goingabout their daily lives, creates an emergentphenomenon that we call society a phenomenon asreal as the emergent pattern of a whirlpool.41

The Nature of Analysis: Mismatched Models

Traditional economic theory has proven inadequate in terms of the two standard criteria for ascientific theory: prediction and explanation. Models in Traditional Economics often usesimplifying and highly restrictive assumptions. Famously, Milton Friedman insisted thatunrealistic assumptions in economic theory do not matter so long as the theories make correctpredictions.42 Such optimism would seem misplaced. Indeed, all mathematical statements areconditional in nature. Assumptions must be appropriate for the purpose of the model, and mustnot affect the answers the model provides for that purpose. In econometrics, statisticalcorrelations do not provide a causal explanation of the phenomena, and data often is not readily

available, or problematic. Paul Ormerod believes that, to be of any value, theories must beconfronted with reality.43 Philip Ball explains further that:

Economic models have been augmented, refined, garlanded, and decorated withbaroque accoutrements. Some of these models now rival those constructed byphysicists in their mathematical sophistication. Yet they still lack theirNewtonian first principles: basic laws on which everyone agrees.44

There also is an uncomfortable feeling that economic models oftentimes simply lose the humanelement in their too-neat equations. We want to believe that economic theory does not regard usas automata and preprogrammed, omniscient computers in order to make their mathematical

37 Ball, Critical Mass, at 211-12.38 Ormerod, Why Most Things Fail, at 125.39 Ormerod, Why Most Things Fail, at 224.40 Malcolm Gladwell, The Tipping Point (2000), at __.41 Beinhocker, The Origin of Wealth, at 450.42 [cite]43 Ormerod, Why Most Things Fail, at xiii.44 Ball, Critical Mass, at 181.


13/77



models work, but instead takes seriously the thousand and one parts of our daily lives thatcannot be reduced to numbers but that makes our lives worth living.45

PRESENTING AROUGH FORMULAFOREMERGENCE

Economic activity fundamentally is about order creation. We organize our world bytransforming energy, matter, and information into the goods and services we want. Bycooperating, specializing, and trading, we can create even more order than we otherwise couldon our own.46

The complex interactions that make up our networked innovation economy are not simple tomodel, and in turn do not lend themselves to simplistic policymaking. This kind of marketoperates as an open, dynamic, and nonlinear system. Emergence Economics is our suggestedumbrella phrase for a rapidly developing field that incorporates a broad set of tools in order tounderstand this type of activity. Eric Beinhocker outlines some of the principles of this approachin his recent book, The Origin of Wealth.47 We have assembled these and other elements into a

rough formula that captures the essence of the economic activity:

AGENTS +NETWORKS +EVOLUTION =EMERGENCE

Agents in this case consist of the full spectrum of economic actorslarge and small businesses,noncommercial organizations, ordinary consumers, individuals with varying motivations,universities that generate foundational research, government officials, and others. These agentsform ad hoc relationships that change over time and interconnect into larger social networks.Through this process, individual agents build on others innovations, and the system overallevolves toward greater productivity. When these dynamics arise, the system develops anemergent structure, generating spontaneous and nonlinear growth. Emergence, then, is what

results from the complex interplay of agents, networks, and evolutionary forces.48

This is not meant to suggest a straightforward linear equation. Each element is its own complexadaptive system, which greatly expands the scale and scope of the resulting emergent behavior.Instead of straightforward emergence (if there is such a thing), we have emergence layered ontop of emergence, many times over. The rough formula thus is an over-simplifiedapproximation intended to illuminate these properties, and should not be misconstrued as afaithful representation of reality. By isolating, we both illustrate, and distort.

Importantly, what we exchange in this system are not just finished goods and services pertraditional economic theory -- but raw ideas and applied technologies, and new means of

productivity. The Net also is the physical infrastructure that supports such growth across manysectors of our economy. Innovation and technology are key elements, because they propagatethrough the network as components of new recipes for economic growth. We will explore morefully in later sections the role of the Internet, and the roles of ideas and innovation, economic

45 Ball, Critical Mass, at 208; 455.46 Beinhocker, The Origin of Wealth, at __.47 Eric Beinhocker, The Origin of Wealth (2006).48 For a far more in-depth treatment of the mathematical equations of evolutionary dynamics, see Martin Nowak,Evolutionary Dynamics (2006).


14/77



growth and Net effects. We will also show how policymakers generally should not attempt toengineer or intrude into these market-based relationships, but still can help keep the system opento productive dynamism. For now, though, we will sketch out the elements of a rough formulafor emergence.

Agents

Any theory of economics must begin with a sound theory of human nature. After all,economics are ultimately made up of people.49 With that overarching premise in mind, wewill briefly examine how Traditional Economics is built on the flimsy and ultimatelyunsupportable premise that human beings are perfect economic agents.

In this paper, we will use the word agent generically to describe humans acting in theirenvironment. To be an agent is to have several different meanings and connotations: as a self-possessed entity, as acting on behalf of others, and in the chemical sense, providing catalyticchange. The term is preferred to either consumer or user, both of which tend to reduce humans

to a one-way relationship of purchasing access to goods, services, or other resources.

Agents are economic actors, and individual nodes in a network. Whether acting as consumers orinvestors, CEOs or government officials, all of us play this central role in the economy. Thecentral insight of economics is that agents respond to incentives.50 Beyond that observation,traditional economic theory assumes that agents have definite characteristics. These assumptionsinclude: agents are modeled collectively, use complex deductive calculations to make decisions,have complete information available for free to gather and process, account for all relevantfactors, face no transaction costs, have perfect freedom to act, make no errors and have no biasesand being perfect have no need for learning or adaptation.51 Under the standard model ofhuman behavior, each of usdisplays perfect rationality, by pursuing our economic self-interest in

carefully calculated ways.52

Economic actors only interact through market prices. AsLeijonhufvud has put it, the usual economic model of human behavior posits incredibly smartpeople in unbelievably simple worlds.53

Each of these assumptions is misplaced. Much well grounded thinking about agents and whatthey do comes from the latest teachings of behavioral psychology and neuroscience. In fact, anew form of economics has emergedbehavioral economicswith the actual human being at itscore. Behavioral economics seeks to right some of the false assumptions that lie at the heart ofTraditional Economics. For example, agents do not possess perfect rationality; instead, at bestthey live with bounded rationality.54 Imperfect and asymmetric information is the rule, rather

49 Beinhocker, The Origin of Wealth, at 115; see alsoShermer, Mind of the Market, at 190.50 Ormerod, Why Most Things Fail, at 63; William Easterly, The Elusive Quest for Growth (2001), at 143.51 Beinhocker, The Origin of Wealth, at 97, 115-139; Ball, Critical Mass, at 204-25.52 Beinhocker at 51; Ormerod, Why Most Things Fail, at 64.53 Axel Leijonhufvud, "Towards a Not-Too-Rational Macroeconomics", Chapter 3 (pp. 39-55) in David Colander(ed.), Beyond Microfoundations: Post Walrasian Macroeconomics.54 Herbert Simon first introduced this concept in the 1950s, but only recently has it begun to influence everydayeconomic thought. Ariel Rubinstein, Modeling Bounded Rationality (2002), at 3; Ball, Critical Mass, at 211-12.

Joseph Stiglitz and George Akerlof, 2001 Nobel Prize winners, have helped further the concept of boundedrationality in economics.


15/77



than the exception, in most high-stakes competition.55 Nor are we homogenous billiard balls orgas molecules, but creatures with different interests, intentions, and biases, all of whichinevitably color whether and how we make economic decisions.56 These aspects of our behaviorstem from the fact that our senses, thoughts, and memory are attuned to the embodied, evolvedenvironment of an early homo sapiens. Survival and procreation, not truth, are the governing

realities that have shaped us.

57

In summary, then, of recent research in the area:

we prefer stories to statistics (relying on anecdotal evidence); we seek to confirm (remembering the hits and forget the misses); we crave causality (underestimating the role of chance and coincidence in life); we misperceive aspects of our world (senses can be deceived); we oversimplify (avoiding analysis paralysis); we have faulty memories (memory is constructive); we hold beliefs based on many external influences (parental, sibling, peer,

educational, social, and cultural); we have framing biases; we rely only on available evidence; we utilize linear processing; we have difficulty accurately calculating risk and probabilities; we can be confused and even paralyzed by having too many options; we compartmentalize our economic behavior; and we have individually varying skills, perspectives, and intuitions.58

In particular, Stanovich notes that while our problem-solving strategies lead us to select regular,deterministic, indication-dependent, functional, and linear processes, the world itself exhibitsirregular, indeterminate, and independent processes.59 We fall for the decision illusions ourminds show us, because we are limited to the tools nature has given us, and thenatural ways inwhich we make decisions is limited by the quality and accuracy of these tools.60

These varying aspects of our behavior obviously directly affect whether, why, and how we makedecisions in the marketplace.61 As just one example, profit-seeking entities can seek actively to

55 Case, Competition, at __.56 Mark Taylor, Confidence Games, at 273.57 To put it more colloquially, one can apply neuroscience metaphorically to the 1992 motion picture A Few GoodMen, where Tom Cruise (as The Brain) asserts I want the truth, and Jack Nicholson (as The World) responds,You cant handle the truth.58 The literature literally teems with excellent treatments of all these well-grounded scientific findings. See, e.g.,George Lakoff and Mark Johnson, Philosophy in the Flesh (1999); Andrew Newberg, Why We Believe What WeBelieve (2006); Tor Norretranders, The User Illusion (1991); Humberto Maturana and Francisco Varela, The Tree ofKnowledge (1998); Gilles Fauconnier and Mark Turner, The Way We Think (2002); Thomas Kida, Dont BelieveEverything You Think (2006); Daniel Schacter, The Seven Sins of Memory (2001); Marc Houser, Moral Minds(2006); Melvin Conner, The Tangled Wing (2003); Robert Fogelin, Walking the Tightrope of Reason (2003); David

J. Linden, The Accidental Mind (2007); Timothy Wilson, Strangers to Ourselves (2002): Cordelia Fine, A Mind ofIts Own (2005); Chris Frith, Making Up The Mind (2007); Mark Blumberg, Basic Instinct: The Genesis of Behavior(2005); Paul Bloom, Descartes Baby (2005); Keith Stanovich, The Robots Rebellion (2005); Nassim Nicholas

Taleb, Fooled by Randomness (2004); Barry Schwartz, The Paradox of Choice (20004); Read Montague, YourBrain Is (Almost) Perfect: How We Make Decisions (2007); Dan Ariely, Predictably Irrational (2008).59 Keith Stanovich, The Robots Rebellion, at __.60 Dan Ariely, Predictably Irrational, at 243.61 See, e.g., Taleb, Fooled by Randomness (humans tend to overestimate causality and underestimate luck); Paul W.


16/77



prey on those constraints as part of the process of selling goods and services, which onlyexacerbates the shaky foundations for an agents market decisions. Others have pointed out thepolitical implications for democratic societies as well.62

Of course, firms are just collectives of individuals, and often act as if possessed of a single mind.

Firms share similar individual characteristics of agents, in particular routinely lacking relevantinformation and possess inherent uncertainty.63 Ormerod states that agents of all types, includingfirms and governments, have very limited capacities to acquire knowledge about the true impactof their strategies on others or of others on them.64 Agents face massive inherent uncertaintyabout the effects of their actions.

Some see the constraints inherent in our human information processing systems as signs ofsignificant and inherent weakness the proverbial glass half empty. However, other researchpoints in the opposite direction: human beings as more capable, multi-faceted, and flexible thanheretofore has been recognized. Again, these fundamental characteristics are not reflected inTraditional Economics. Among the key findings:

we have a variety of motivations, including non-economic ones; we utilize not just reason but imagination, intuition, and insight as the foundations for

creative thinking; we often know more than the official producers; we can understand and even transcend our constraints; we are wired to engage in market-exchange calculations; we are altruistic, cooperative, and sharing creatures; we can use intelligent action to tip the world in certain directions; we use induction (pattern recognition), as well as deduction (the scientific method);

and we possess important traits as autonomous entities interacting to carry out particular

tasks.65

Most importantly, human beings have an inherent ability to learn, to adapt, to change and togrow. We evolved the adaptation of adaptability.66 Our brains have been created with built-inplasticity, so that they are malleable and open to conscious change from new experiences and

Glimcher, Decisions, Uncertainty and the Brian: The Science of Neuroeconomics (2003); Richard Restak, TheNaked Brain (2006). Recent books have also begun to apply lessons from biology and complexity science to themanagement of large organizations. See, e.g., John Henry Clippinger III, ed., The Biology of Business (1999)(collection of essays explaining the Complex Adaptive System of management); Robert Axelrod and MichaelCohen, Harnessing Complexity (2000).62 Bryan Caplan, The Myth of the Rational Voter (2007); Drew Westen, The Political Brain (2007).63 Ormerod, Why Most Things Fail, at 21-35.64 Ormerod, Why Most Things Fail, at 221.65 See, e.g., Y ochai Benkler, The Wealth of Networks; Howard Gardner, Changing Minds (2004); Richard Ogle,Smart World (2007); Eric Von Hippel, Democratizing Innovation (2005); Nassim Nicholas Taleb, The Black Swan(2007); Michael Shermer, The Mind of the Market (2007); James Odell, Agents and Complex Systems (2002),found at: http://www.jot.fm/issues/issue_2002_07/column3/; Malcolm Gladwell, The Tipping Point, at 259; MattRidley, The Origins of Virtue (1997). In Benklers memorable words, it turns out that we are not intellectuallemmings. Benkler, The Wealth of Networks, at 466.66 Shermer, Mind of the Market, at 190.


17/77



new learnings.67 We are adaptive agents (or more precisely, agents capable of adaptation) in anever-evolving landscape. Nor are we the selfish automatons that Traditional Economicspresupposes.68 Stanovich further insists that we can use our rational self-determination to gaincontrol over our mismatched genetic and cultural programming,69 while Donald surmises that wecan take advantage of our hybrid brain/cultural mind to break free from our evolutionary

heritage.

70

As populations of agents, we can learn from each other, share new ideas andinnovations, and serve as a fertile environment for growth.71 We also have recourse to a vastarray of culturally and socially embedded idea-spaces that populate our extended minds.72

Another point is worth stressing here: the traditional focus on the single individual, standingalone in her perfect wisdom and forethought, ignores growing evidence that large groups ofpeople are better than the few agents, or the one, at solving problems, reaching decisions,predicting the future, and fostering innovation.73 Philip Ball puts it well:

One of the features of collective behavior arising from local interactions is that itbecomes impossible to deduce the global state of a system purely by inspectingthe characteristics of its individual components. This is physical sciences mostimportant message to social science: do not be tempted too readily intoextrapolating from the psychology of the individual the behavior of the group.74

While we will return to this concept at a later point, for now the crucial takeaway is thatTraditional Economics has little to say about collective intelligence operating in the marketplace.

In short, evolutionary theory, neuroscience, and behavioral studies together show that humansare both more limited, and more limitless, than Traditional Economics has assumed. We areeminently fallible, yet highly adaptable, agents. Any well-grounded economic theory must takenuanced account of both the half-empty and half-full views of economic agents. Neoclassicaleconomic theories fail this fundamental test.

Networks

Of course, constrained yet adaptable agents (normal human beings) do not exist in a vacuum.The full productive potential of agents comes from their interactions with each other, whichfacilitate sharing of information and effort. Any particular agent may have a link to several otheragents, which in turn link to others. These links consist of lines of communication, commontasks, market agreements, or any number of other relationships.

67 Jeffrey Schwartz and Sharon Begley, The Mind and the Brain (2002).68 In an often-overlooked work, Adam Smith declares: How selfish soever man may be supposed, there areevidently some principles in his nature, which interest him in the fortunes of others, and render their happinessnecessary to him, though he derives nothing from it, except the pleasure of seeing it. Adam Smith, The Theory ofMoral Sentiments (1759), at __.69 Stanovich, The Robots Rebellion, at 95-171.70 Merlin Donald, A Mind So Rare (2001).71 Axelrod and Cohen, Harnessing Complexity, at 5.72 Ogle, Smart World, at __.73 James Surowiecki, The Wisdom of Crowds (2004); Howard Rheingold, Smart Mobs (2002); Clay Shirky, HereComes Everyone (2008).74 Ball, Critical Mass, at 298-98.


18/77



In Traditional Economics, agents only interact indirectly, through static and closed marketmechanisms.75 As a result, many of the connections within the economy are downplayed, oreven ignored.76 Reality is a bit more complex than that. In a dynamic system, relationships arebound to change over time. The true value of an agent is affected, and often greatly enhanced,by links to other agents. It is the structure of the connections between the component parts that

gives systems of people their distinctive and characteristic features.

77

When viewed as a whole,human systems shows themselves to be complex sets of relationships that can be understoodfully neither from the perspective of the individual agents, nor the system as a whole.

The economy is best conceptualized and analyzed as a connected system, a network of individualagents. In the language of network science, the agents are nodes and the links are edges.Thus, networks are interactions of nodes, and groups of edges between and among nodes. Inparticular, the interactions of agents (whether individuals, firms, or governments) includeinherent elements of unpredictability, and helps create complexity in the overall system.Networks both define, and are defined by, these interactions.78

The characteristics of networks can sometimes be counterintuitive, but when understood they canbe extremely powerful. The field of network science explores how networks form, and attemptsto explain why certain dynamics arise in networks that do not appear in more static, linearsystems.79 In fact, there is a growing consensus that common structures, growth patterns, andcollective behaviors will arise in networks composed of very different kinds of elements andlinkages.80

As mentioned previously, physicists and biologists for decades have been studying complexadaptive systems. These are open systems of interacting agents that adapt to each other and theenvironment. Examples of CASs include neurons in the brain, immune systems, biologicalecosystems, and the Internet. Economies too are a type of complex adaptive system. Suchcomplex systems may be understood as energy flow structures organized by thermodynamic

principles.81 Economic and social systems are essentially dynamic, and not static.82 Somehave termed it the econosphere -- the economy as a dynamic, evolving system83.

In many systems, individual actors end up having indirect positive effect on others. Economistscall these effects positive externalities, and often discuss the benefits that accrue to others asspillovers.84 For example, I may invent a new method for scanning bar codes that yields me

75 Beinhocker, The Origin of Wealth, at 97, 141-159.76 Ormerod, Why Most Things Fail, at 146.77 Ormerod, Why Most Things Fail, at 173.78 The network is the dominant pattern of the new digital economy. W. Brian Arthur, Myths and Realities of theHigh-Tech Economy, talk given at Credit Suisse First Boston Thought Leader Forum, September 10, 2000, at 1.79 Duncan Watts, Six Degrees (2003); Steven Johnson, Emergence (2001); Albert-Laszlo Barabasi, Linked (2002).80 Katherine Strandburg et al, Law and the Science of Networks, Berkley Technology Law Journal, 21:4, 1293, 1301(2006).81 Eric Schneider and Dorion Sagan, Into the Cool: Energy Flow, Thermodynamics, and Life (2005), at 293.82 Paul Ormerod, Why Most Things Fail, at 18-21, 50-51.83 Science writer Philip Ball criticizes the practice of using the term complexity science to explain aspects ofhuman behavior. He relies instead on the concept of a science of collective behavior, and sees the market lawsemerging from the ordinary (but still unpredictable) push and pull of trade. Ball, Critical Mass, at 5-6; 179-180.84 Some prefer to call these externalities demand side economies of scale. See Mark Cooper, Wifi, Wikis, andOpen Source, at 133.


19/77



great profit, but you might adopt or adapt this technology to your own benefit (provided that thelaw allows). In complex networks, these benefits flow more freely than in disconnected islands.The type of externalities referred to as network effects arise only in networks. In this case,each new node added to the network creates added value for the existing nodes. The classic caseis the telephone network, in which a globally interconnected system is substantially more

valuable to all than a regional or locally delimited system.

A more recent example is the digital network Ethernet standard. The more people that ownedEthernet equipment, the more useful the network which eventually helped catalyze theexplosion of consumer Internet use. The presence of externalities means that a great deal ofwhat happens in a network, and the value that is created, comes from and flows to other nodes.It also means that the total value created is greater than what each node can create or capture inisolation. In other words, a network becomes more valuable to its users as it grows.85 We usethe term Net effects later in this paper to refer to a diversity of presumed externalities that infact arise internally from the complex network itself.

Network formation theory looks at networks as endogenous factors that both produce and areproduced by a collection of interactions.86 There are two broad classes of hownetworks form:random formation, from graph theory (as formulated by Spulber and Yoo),87 and strategicformation of individual, self-interested agents, from game theory (as formulated by Werbach).88

To begin with, networks have a tendency to expand slowly and then exhibit explosive growth asindividual networks interconnect. Positive externalities accelerate this activity, because thesehighly interconnected networks represent considerably more value, and the effects of each newnode feeds back into the system. However, these types of phase transitionsabrupt jumpsfrom one state of connectedness to anothercan also work in the reverse direction.89 If thosewho control particularly central nodes, edges, or clusters see benefit in restricting use of thoseassets, they can exponentially dampen the growth of the network as a whole.90

Another feature of networks is that they can help reduce transaction costs of finding andnegotiating interactions with partners. This is true both of literal networks, like the Internet, andof social or market relationships. An isolated node would have to generate its own value ornegotiate with others to obtain what it needed. Traditionally, the presence of these transactioncosts has been used to explain why firms are created.91 By bringing many entities togetherunder a single umbrella, an organization can limit the transaction costs required. In complexnetworks, these units need not be limited to literal firms, and the multitude of links can reducetransaction costs in more dynamic fashion.

85 Here is another instance where a tenet of Traditional Economics most markets are characterized by decliningand eventually negative returns to scale -- does not necessarily comport with reality.86 Werbach,The Centripetal Network, at __.87 Daniel Spulber and Christopher Y oo, On the Regulation of Networks as Complex Systems: A Graph TheoryApproach, [cite.]88 Werbach,The Centripetal Network, at __.89 Werbach,The Centripetal Network, at __.90 While such restrictions can appear to make rational sense from the perspective of one agent, another agent withbetter understanding of the greater dynamics at work likely will find a way to avoid such counter-productivebehavior, while also capturing more value than an isolationist approach would yield.91 R.H. Coase,The Nature of the Firm(1937).


20/77



Complex real world networks exhibit three other kinds of behavior worth noting here. Smallworld behavior states that the diameter of a network (the average number of links between anytwo nodes) tends to grow much more slowly than the number of nodes.92 This means that arelatively small number of hops is necessary to connect any two nodes in the network. In otherwords, a small worlds network is relatively tightly connected.93 Scale-free dynamicsstates that

some nodes are vastly more connected than others, so that additional links are more likely toconnect to nodes that are already well connected. This behavior explains the so-called rich getricher effect, where preferential attachment by new users is a real element of networks.94Finally, self-organized criticality and critical points refer to a networks state of precariousstability, where one of several paths is imminently possible.95 Taken together, these threecharacteristics provide important insights on how and why complex networks like economiesbehave the way they do.

Evolution

Traditional Economics has no explicit mechanisms for explaining endogenous novelty (within

the system), or agents who learn and adapt, or sudden growth in complexity.96 By contrast,Emergence Economics uses the universal algorithm of evolution as the basis for much of itsanalysis.

Ultimately, economics is a biological science. It is the study of how humans choose. Thatchoice is inescapably a biological process.97 Ilya Prigogine explains that we live in anevolutionary universe, where the laws of nature no longer deal with certitudes but possibilities,and irregular, chaotic motions constitutes the very foundation of macroscopic systems.98 Theeconomy is one such macroscopic system, and, as we have seen, specifically a complex adaptivesystem. As such, evolution becomes the ideal algorithm for creating value within that system, aniterative process of experimentation by agents that includes first differentiation, then selection,

and finally amplification of things that work. Hayek, another transitional figure in economics,was one of the first to gain the insight that markets involve the evolutionary formation of suchhighly complex self-maintaining order.99

The Thr ee Stage Process

Evolution is the universal algorithm for change in biological systems, and now has beenidentified as operating within economic systems as well.100 Natural selection is simply adescription of certain evolutionary processes initiated by agents.101 In economic systems, one

92 Duncan Watts, Small Worlds (1999); Mark Buchanan, Nexus (2002).93 Strandburg et al,Law and the Science of Networks, at 1305.94 Strandburg et al,Law and the Science of Networks, at 1308-09.95 Ball, Critical Mass, at 227-241.96 Beinhocker, The Origin of Wealth, at 97, 187-217.97 Glimcher, Decisions, Uncertainty, and the Brain, at 336.98 Ilya Prigogine, The End of Certainty (1996), at 155.99 Hayek, The Fatal Conceit: The Errors of Socialism (1988), at __.100 Geerat Vermeij, Nature: An Economic History, at 43-58.101 Shermer, Mind of the Market, at 41. Taleb clarifies that natural selection does not in fact do anything it isnot a mechanism or causal agency. In reality, the differential selection of a trait, or an adaptation, is a consequence


21/77



can usefully think of the process of natural selection as comprised of three interrelated stages:differentiation, selection, and amplification. The first step of evolution is differentiation, inwhich agents identify and propose various possible approaches. Next, through observation andaction, these agents sort through the variation to find what works and what does not, and selectthe most fit solutions. Finally, the agents share anditerate on the most successful approaches,

throwing out the others and amplifying the effects.

102

In other words, natural selection bothweeds out what fails to work, and weeds in what does.103

So adaptation is the formation, and continual testing, of hypotheses about the environment.104This same evolutionary formula lies at the heart of the market process. Agents, acting asselectors, pick and choose which products and services and other transactions they want toengage in, and other agents respond accordingly. They use the infrastructure of the network asthe environment to evolve both their practices and the structure of the network itself. Out of thisastonishingly complex series of moves, an ordered market system evolves.

Evolution allows for experimentation with a variety of solutions to problems, means ofinnovation, and shared experience. Many problems we encounter are complex, and lack clearideal paths to a solution. Scientific discovery has long exhibited this hit-or-misscharacteristic,105 and technological breakthroughs similarly can come from unexpecteddirections. As Daniel Dennett puts it, evolution finds needles of good design in haystacks ofpossibility. Evolution discovers design, through trial and error, acting as The GreatTinkerer.106 Chance and accident also play a significant role in evolution,107 as well as simpleluck.108 If we assume a particular design space in which agents experiment, they can adaptsuccessful designs by continuing to iterate on what proves useful, and eventually converge onone or more fitness functions.109 The environment is the design space of evolution; the market the econosphere or marketspace is the design space of economics. People use this

of the functional effects it produces in relation to the survival and reproductive success of an organism in a givenenvironment. It is these functional effects that are ultimately responsible for the trans-generational continuities andchanges in nature. Taleb, Black Swan, at 16.102 For a more complete overview of the basics of evolution in a networked economy, see Beinhocker, The Originof Wealth, at 213-216.103 Taleb, Black Swan, at 17. Other analysts employ a somewhat different schema for the evolutionary process. Forexample, authors Axelrod and Cohen divide up the evolutionary algorithm into Variation (the raw material ofadaptation), Interaction (between agents and populations of agents) and Selection (to promote adaptation). RobertAxelrod and Michael Cohen, Harnessing Complexity, at 32-151. They explain that harnessing complexity refersto changing the structure of a complex system in order to increase some measure of performance. Id. WilliamWallace talks about technology creating disruptions to the economy that trigger the FROCA process (Frontier,Release, Overexploited, Crash, Adaptation). William Wallace, Techno-Cultural Evolution (2006), at 7. By aninteresting reverse analysis, Geerat Vermeij shows how processes common to all economic systems competition,cooperation, adaptation, and feedback in turn also govern evolution. Geerat J. Vermeej, Nature: An EconomicHistory (2004).104 Vermeij, Nature, at 55.105 [See Thomas Kuhn, others].106 Daniel Dennett, Darwins Dangerous Idea (1996), at __.107 Stephen Jay Gould, Wonderful Life (1989), at 285, 288 (Our own evolution is a joy and a wonder because sucha curious chain of events would probably never happen again, but having occurred, makes eminent sense. Themodern order is largely a product of contingency.).108 The reason free markets work is that they allow people to be lucky, thanks to aggressive trial and error, not bygiving rewards or incentives for skills. The strategy, then, is to tinker as much as possible. Taleb, Black Swan, atxxi.109 Beinhocker, The Origin of Wealth, at __.


22/77



design space to purchase, sell, and barter the goods and services best suited to meet their uniqueneeds and desires.110

Fitness is the measure of the potential for value creation; it is a contingent concept, premised onthe challenges and opportunities of a particular environment. Fitness is measured by the

capacity to connect and interrelate effectively and creatively.

111

Ormerod explains that if weincrease the fitness threshold at which agents become extinct, we are making it more difficult forthem to survive, and if we reduce it, we are making it easier. We can readily think of this ascorresponding to more and less competitive environments, respectively.112 Because thesolution space of fitness is complex and often changing, this is not a linear process. Instead, itis continuous innovation that takes place in parallel and works by building shared knowledge thatfeeds back into the system.

Although the three-stage formulation of the evolutionary process sounds simple, there are severalchallenges to successfully employing it. One problem is the sheer number of possible formulas.To try each one can prove to be impractically complex, or the partnerships required canintroduce insurmountable transaction costs. To complicate further the process, it can be difficultto discern whether an idea will prove fruitful until several iterations are made or until acomplementary approach is developed. In this case, it helps to encourage a plethora ofexperimentation with minimal barriers to cross-pollination. Finally, successful evolution canonly take place when experimenters overcome the social tendency of path dependence, inwhich agents simply do things as they have always been done.

Ormerods Iron Law of Failure further explains that it is failure, not success, which is thedistinguishing feature of corporate life.113 Most firms fail. On average, more than 10 percent ofall economically active firms in the United States become extinct each year, with roughly thesame number of new firms added back to the market.114 As part of this process, weaker firmsare replaced by firms with higher levels of fitness to the existing environment.115 Traditional

economic theory simply ignores this widespread existence of corporate failure.116 As biologicalevolution relies on accident mutation as the basis for potential change, so do entities in the

110 Shermer, Mind of the Market, at 8.111 Taylor, The Moment of Complexity, at 197.112 Ormerod, Why Most Things Fail, at 230. Others note that strong selection pressure amplifies the success of thebest agent while diminishing overall variety in the system, while weaker selection pressure provides more varietybut sacrifices some agent fitness. Harnessing Complexity, at 129-130.113 Ormerod, Why Most Things Fail, at 12.114 Ormerod, Why Most Things Fail, at 180.115 Taleb claims that the concept of evolutionary fitness is overstated, and that evolution ultimately is a series offlukes, some good, some bad. The fools, the Casanovas, and the blind risk takers are often the ones who win in theshort term. Taleb, Black Swan, at 116-17. Some companies survive simply because they were the lucky ones.

Taleb insists we should love free markets because operators in them can be as incompetent as they wish. Id. at181.116 Ormerod, Why Most Things Fail, at 17-35. Ball agrees that many economic theories of the firm fail toacknowledge that most firms are ephemeral. Ball, Critical Mass, at 267. The larger lesson is that, for selection tooccur, the system needs superfecundity more designs than the environment can support which thus createscompetition. In biology, there are more potential organisms than any ecosystem can support. The sameundoubtedly is true for the market, which helps explain Ormerods Iron Law of Failure.


23/77



economic environment oftenprosper, or fail, due to the exigencies of a particular environment in other words, fickle fortune.117

Two Types of Technology

Evolution operates on two broad types of technologies, what Beinhocker refers to as PhysicalTechnologies and Social Technologies.118 Physical Technologies are means or recipes forproducing objects or ideas; they consist of specifications, instructions, shareable practices, andother ways of transforming materials in order to serve a goal. These technologies have amodular, building-block character, of components plus architecture, and instill order in thephysical realm. Social Technologies, on the other hand, are methods and designs for organizingpeople in service of a goal, and instill order in the social realm. This might consist of a particularteam structure or collaborative relationship. The modern day corporation is seen by some as anenabling technology in its own right, crucial to economic development.119

In reality, the two types of technologies evolve in relation to each other, and with concrete

business designs (referred to in Beinhockers work as Business Plans) that incorporate one orboth.120 A software company might find that one specific software development toolkit makesits work easier, and that small working groups of engineers further improves productivity.Physical Technologies can enable Social Technologies, and vice-versa. Each type of technologyconstitutes an evolution of modular ideas that has the potential to be plugged into otherscenarios. As with firms, technologies are subject to their own law of failure in the market.121The long-term power of these successful technologies lies in their capacity to be shared and re-used.

As we have seen, biological ecosystems provide a powerful analogy and insights to thefunctioning of business networks. Under one approach, companies work to connect a large and

distributed network of companies to their customers, and provide platforms that other firmscan leverage to increase productivity, enhance stability, and spur innovation.122 The keystoneis a pattern of behavior that improves the performance of an ecosystem and, in so doing,

117 Again, Taleb finds luck to be the grand equalizer in a free market, because almost everyone can benefit from it,and it is far more egalitarian than even intelligence. Randomness reshuffles societys cards, sometimes knockingdown the big guy. Taleb, Black Swan, at 222. Ormerod observes as one example the success of MicrosoftsWindows operating system (OS), which was far more the result of a series of accidents than of a far-sighted,planned strategy. Ormerod, Why Most Things Fail, at 122; 123-124.118 Beinhocker, The Origin of Wealth, at __. Others see the proper unit of selection in the market as occupations, ormaking a living, rather than technology. Vermeij, Nature: An Economic History, at 44.119 John Micklethwait and Adrian Wooldridge, The Company (2003), at xxi. The company has flourished inmodern markets because capital can be pooled for investment, investor risk is spread, transaction costs are reduced,and effective management structures are imposed on large organizations. Id.; see alsoBall, Critical Mass, at 250-54.120 Vermeij observes that in organisms, technology is part of the body; in people, it is an extension mechanical,intellectual, and cultural that we design and that, at least figuratively speaking, takes on a life of its own. In bothcases, technology evolves; in organisms it does so largely through natural selection, in humans by engineering andmarket forces. Vermeij, Nature: An Economic History, at 47. Kurzweil has commented that technology is thecontinuation of evolution by other means. Quoted in Taylor, The Moment of Complexity, at 221.121 As Romer has remarked, there are many more dead ends out there than there are useful things to discover.Paul Romer interview with Ronald Bailey, Reason Online, December 2001.122 Marco Iansiti and Ray Levien, The Keystone Advantage (2004), at __.


24/77



improves individual performance.123 Just as keystone species in nature play central roles intheir ecosystems, companies such as Wal-Mart, Microsoft, and Li and Fung deploy keystonestrategies, using effective collaboration to actively shape and regulate the workings of theirbusiness ecosystems.124

Losing Ones Balance

All of this flies in the face of traditional economic notions of linear progression and naturalequilibrium. Neoclassical economists seek a single optimal balance for a particular market, andsee growth as a smooth trajectory of improved efficiency and increased output. Our morecomplex view of the process acknowledges that there are several possible peaks of highproductivity that operate in different ways, and that it is possible to arrive at those peaks viadifferent fitness functions. Indeed, just when one peak has reached its maximum utility (say,bamboo-based light bulb filament), an entirely different approach might offer a far better fit(such as tungsten-based light bulb filament).

The notion of fitness implies that combined Physical Technologies (PTs) and SocialTechnologies (STs) are used by agents to navigate a market landscape of possible growthtrajectories like a map of mountains. Agents combine PTs and STs into a Business Plan,according to various strategies. As one approach reaches its limit or a peak, one might say thatan equilibrium of sorts has been reached -- but only until it is upset by a different approachmaking use of a different combination. This leads to a punctuated equilibrium that is disruptedby keystone technologies.

differentiation

selectionamplification

BP

PTST

Ultimately, no one company can hope to out-innovate the market. An ecosystem tends to beat aproduct (perhaps even something as innovative as the iPod) because its collective of competitorscan explore and innovate and invest in many more ideas than any single company can muster.125Beinhocker observes that in evolutionary systems competitive advantage does not exist; there isonly a never-ending race to create new sources of temporary advantage. The bottom line isevolution is cleverer than you are.126

123 Marco Iansiti and Ray Levien, The Keystone Advantage, at 113-115.124 Iansiti and Levien, The Keystone Advantage, at __.125 John Sviokla, Fast Company, at __.126 Beinhocker, The Origin of Wealth, at __.


25/77



Emergence of Networks and Growth

Decades of research shows that the economic system is a complex adaptive system, where microinteractions of agents lead to macro structures and patterns. In other words, more isdifferent.127

Emergence is not some mystical force that magically comes into being when agentscollaborate.128 Emergent properties are aspects of a system not otherwise exhibited by thecomponent parts. They are macro-level features of a system that arise from interactions amongthe systems micro-level components, bringing forth novel behavior.129 The brain is an example:the single neuron has no consciousness, but a network of neurons brings forth, say, the smell of arose. Similarly, when agents interact through networks, they evolve their ways of doing workand discover new techniques. Out of this combined activity, a spontaneous structure emerges.Without any centralized control, emergent properties take shape based on agent relationships andthe conditions in the overall environment. Thus, emergence stems from behavior of agents,system structures, and exogenous inputs.130

Emergent systems are often described as being organism-like in the sense that they areconstantly growing and adapting. Each agent follows localized rules and motivations, but theend result is additive and interdependent. Analogies drawn from biology include the ant colony.Ants follow basic rules for seeking food, emitting pheromones to leave a trail to the food theyfind, and following other ants pheromone trails to make their way to food and back to thecolony. These characteristics appear in many human systems. James Odell notes that, with thestock market, thousands of agents act independently to buy and sell shares of particular stocksand bonds. Yet from this independent behavior, an organism-like product called the stock marketemerges.131 Much of the development of cities similarly derives from the bottom-up.132

Emergent systems have no single ideal structure. They exist in an ever-changing environment,

and consist of complex interactions that continuously reshape their internal relationships. BrianArthur notes that our subjective beliefs constitute the very DNA of the market, and so co-evolve, arise, decay, change, mutually reinforce, and mutually negate. The market emergesfrom subjectivity and falls back into subjectivity.133 The many independent actions of agentsunify, but they do not necessarily work toward one particular structure or equilibrium. For

127 See, e.g., Klaus Mainzer, Thinking in Complexity (4th ed. 2003); Mark Buchanan, Ubiquity (2001); TerryBossomaier and David Green, Patterns in the Sand (1998); John Holland, Emergence (1998); John Holland, HiddenOrder (1995); Roger Lewin, Complexity (1992); Scott Camazine et al, Self-Organization in Biological Systems(2001).128 Steven Johnson, Emergence (2001), at 116.129 Tom De Wolf and Tom Holvret, Emergence vs. Self-Organization (2005), at 1. Characteristics of emergentsystems include micro-macro effects, radial novelty, coherence, interacting parts, dynamical, decentralized control,bi-directional links between the macro- and micro-levels, and robustness and flexibility. Id. at 3-5.130 Beinhocker, The Origin of Wealth, at 185.131 James Odell, Agents and Complex Systems (2003).132 Citizens solve local problems, combining resources and expertise in the form of new technologies. Steven

Johnson describes how early cities evolved around new farming mechanisms, with urban emergence intensifying asfossil fuel technologies were developed. And with that new flow of energy, new kinds of cities emerged: thefactory towns of Manchester and Leeds, and the great metropolitan superorganisms of London, Paris, and New

York. Steven Johnson, Emergence (2001), at 113.133 W. Brian Arthur, The End of Certainty in Economics, reprinted in The Biology of Business (1999), at 6.


26/77



example, emergent systems can be robust to change, and they can be far better at evolvingtoward efficiency than top-down systems. On the other hand, emergent structures can fall apartwhen their basic conditions are altered in such a way that they work against the health of thesystem as a whole. If the ants stop leaving pheromone trails, they can no longer cooperativelyfeed the colony. If corrupt stockbrokers are allowed to manipulate unethically the system, the

complex structure of price signals falls apart. If cities are saddled with stagnant industries, theirgrowth falters and their economies can crumble. The line between emergence-fostering actionsand emergence-stifling actions sometimes can be difficult to discern.

Agents actions in turn affect the other agents, setting off both positive and negative feedbackloops. Beinhocker uses the helpful metaphor of adjusting shower temperatures. The delaybetween adjusting the knob and the change in temperature means that one is likely to over-shoot,oscillating back and forth until finally settling on the right temperature.134 But this is a simplecase with a single agent. In a recent study, an economist and a physicist sought to understandwhat happens in youth hostels where many showers share the scarce market for hot water.They found that:

Tuning one's shower in some hotels may turn into a challen

Documents

Emergence Economics